Curious if they'd actually run it, would be the cleanest test of "novel" vs "memorized" so far.
LLMS
-
Models Overthinking Reduces Performance in Benchmarks
By
–
Max thinking doing worse than non-thinking is the funniest part of this benchmark, models really do overthink now.
-
Ollama Qwen 3.6 discovers same vulnerability as Codex-5.4
By
–
Ollama + Qwen 3.6 pulling the same vuln as Codex-5.4 would be huge, keeps the open stack honest.
-
Deploying Hermes: Running Better AI Models Efficiently
By
–
You need to get a Hermes up and running and get a better model to run them.
-

Google’s Reasoning-Driven Synthetic Data Generation Approach
By
–
“Reasoning-Driven Synthetic Data Generation and Evaluation” This paper from Google gave us some insights on how frontier labs currently generate synthetic data. The key is that they treat synthetic data generation as distribution design rather than prompt design, as this builds
-
Major AI Companies Overhaul Apps and Release New Models
By
–
This felt like the week every big AI company decided to overhaul their apps. Here’s all the major news: – @AnthropicAI released Claude Opus 4.7
– Anthropic redesigned Claude Code app
– Anthropic launched Claude Design
– @OpenAI updated the Codex app
– OpenAI launched -

Anthropic and OpenAI advance remarkable AI research and innovation
By
–
Amazing. This was only a dream 5 years ago. Both @AnthropicAI and @OpenAI teams are doing remarkable AI research and innovation. Kudos to them 👏👏 https://t.co/CHaLZWS1Iw
— Nando de Freitas (@NandoDF) 17 avril 2026Amazing. This was only a dream 5 years ago. Both @AnthropicAI and @OpenAI teams are doing remarkable AI research and innovation. Kudos to them
-
Open Source Security and LLMs Impact on Code Safety
By
–
Maybe? I've long argued open source isn't inherently less buggy/more secure but the *process* around it can make it easier to secure. Either way, LLMs are changing how we should think about securing code…
-

Unity AI Gateway Introduces Coding Agent Support Framework
By
–
The era of production-ready coding agents is here, but so is the risk of coding agent sprawl. Today, we’re introducing Coding Agent Support in Unity AI Gateway to bring these tools under a unified governance layer: – Centralized Governance across coding agents, LLM
-
LLMs and Coding Agents Excel Beyond Greenfield Development
By
–
Is there still a widespread belief that LLMs and coding agents are good for greenfield development but don't help for maintaining large existing codebases? I don't think that idea holds up any more